This dataset was recorded in a quiet office environment, with 50 speakers participating, including 23 males and 27 females. All speakers involved in the recordings were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover information such as numbers, dates, times, and personal names.